A study published on Thursday (16) in Natural medicine Artificial intelligence (AI) is able to detect up to 13% more breast cancer cases than doctors, according to Folha des Paulo.
Research has shown that AI can significantly reduce the number of diseases in their early stages that go undetected. It is estimated that 20% or more of cases go undetected by current screening tests without the use of technology.
Read more:
The study, conducted by Imperial College London (England), also showed how AI could help in the fight against cancer, being able to detect errors or signs that are difficult to interpret and that we end up ignoring.
Our study shows that the use of AI can serve as an effective safety net, a tool to prevent more subtle signs of cancer from going unnoticed.
Ben Glucker, co-author of the study and professor of machine learning for imaging at Imperial College London
Research work
- The study used Mia, an artificial intelligence tool developed by Khairon Medical Technologies, from the United Kingdom, which specializes in medical diagnosis using artificial intelligence.
- The researchers conducted an analysis of data collected from 25,000 women who underwent breast cancer screening tests in Hungary between 2021 and 2023.
- There were three phases of the study, and in each, radiologists used AI in different ways;
- The three groups demonstrated improvements in disease detection rates of 5%, 10%, and 13%, respectively, compared with standard readings performed by at least two radiologists;
- Most of the additional cancers discovered were invasive, that is, able to spread to other parts of the body (metastasis).
Seeing that the use of AI can significantly reduce the rate of undetected cancers in breast screening is a huge boost to our mission of transforming cancer care with this resource.
Ben Glucker, co-author of the study and professor of machine learning for imaging at Imperial College London
“Benign Artificial Intelligence”
Research shows that AI is able to improve accuracy when it comes to identifying malignant tissue and even speed up the process.
Studies conducted in Sweden and published at the end of August indicated that mammograms analyzed with the help of artificial intelligence reached a cancer detection rate similar to a standard human double-reading.
Catherine Halliday, president of the UK’s Royal College of Radiologists, said the research involving Hungarian data was a “promising example of how AI can be used to speed up diagnosis and treatment” in the NHS.
The results highlight the potential of artificial intelligence to improve the accuracy of mammogram interpretations and support clinical decision-making. The study underscores the complementary nature of AI and radiologists, and envisions a collaborative future that combines the strengths of both.
Catherine Halliday, President of the Royal College of Radiologists in the United Kingdom
According to researchers, Mia can save up to 45% of the time you spend reading breast cancer screening tests. According to Khairon, the owner of the tool, Mia has been tested in 16 hospitals in the United Kingdom and is being applied in the United States of America.
For researchers, it is extremely important to continue research to expand and deepen the way in which AI can be used to detect disease.
They noted that areas to be prioritized include obtaining results from other countries and using other purpose-driven AI systems similar to Mia, as well as surveillance to identify the emergence of more cancer cases in their study group.
